Antidistillation Sampling
Authors
Yash Savani, Asher Trockman, Zhili Feng, Yixuan Even Xu, Avi Schwarzschild, Alexander Robey, Marc Finzi, J. Zico Kolter
Categories
Abstract
Frontier models that generate extended reasoning traces inadvertently produce rich token sequences that can facilitate model distillation. Recognizing this vulnerability, model owners may seek sampling strategies that limit the effectiveness of distillation without compromising model performance. Antidistillation sampling provides exactly this capability. By strategically modifying a model's next-token probability distribution, antidistillation sampling poisons reasoning traces, rendering them significantly less effective for distillation while preserving the model's practical utility. For further details, see https://antidistillation.com.
Antidistillation Sampling
Categories
Abstract
Frontier models that generate extended reasoning traces inadvertently produce rich token sequences that can facilitate model distillation. Recognizing this vulnerability, model owners may seek sampling strategies that limit the effectiveness of distillation without compromising model performance. Antidistillation sampling provides exactly this capability. By strategically modifying a model's next-token probability distribution, antidistillation sampling poisons reasoning traces, rendering them significantly less effective for distillation while preserving the model's practical utility. For further details, see https://antidistillation.com.
Authors
Yash Savani, Asher Trockman, Zhili Feng et al. (+5 more)
Click to preview the PDF directly in your browser